Welcome to Alpha Odds
Your comprehensive guide to prediction market trading
Master Prediction Markets
Learn everything from platform mechanics to advanced quantitative strategies. This knowledge base contains researched, verified information to help you trade smarter.
Choose Your Learning Path
Beginner
New to prediction markets? Start with the fundamentals.
- How Polymarket Works
- Understanding Odds
- Your First Trade
Intermediate
Understand market dynamics and develop strategies.
- Market Efficiency
- Liquidity Analysis
- Trading Strategies
Advanced
Master quantitative methods and algorithmic trading.
- Kelly Criterion
- Bayesian Methods
- ML Forecasting
Key Platform Facts
Quick Reference
Essential information at a glance
API Base URLs
| Service | URL |
|---|---|
| CLOB REST | https://clob.polymarket.com |
| Gamma REST | https://gamma-api.polymarket.com |
| Data REST | https://data-api.polymarket.com |
| WebSocket | wss://ws-subscriptions-clob.polymarket.com/ws/ |
Contract Addresses (Polygon)
| Contract | Address |
|---|---|
| USDC | 0x2791Bca1f2de4661ED88A30C99A7a9449Aa84174 |
| CTF Core | 0x4D97DCd97eC945f40cF65F87097ACe5EA0476045 |
| CTF Exchange | 0x4bFb41d5B3570DeFd03C39a9A4D8dE6Bd8B8982E |
| NegRisk Exchange | 0xC5d563A36AE78145C45a50134d48A1215220f80a |
API Rate Limits
| Endpoint | Limit/10s | Per Second |
|---|---|---|
| GET /book, /price | 1,500 | 150 |
| POST /order | 3,500 burst | 350 |
| DELETE /order | 3,000 burst | 300 |
| Gamma General | 4,000 | 400 |
Fee Structure
| Market Type | Maker | Taker |
|---|---|---|
| Standard markets | 0% | 0% |
| 15-min crypto | 0% | ~1-3% |
| Polymarket US | 0% | 0.10% |
Essential Formulas
f* = (p - market_price) / (1 - market_price)
Optimal position size when p > market_price
BS = (1/N) * sum((p - o)^2)
Lower is better (0 = perfect)
EV = (p * payout) - ((1-p) * stake)
Only trade when EV > 0
Imbalance = (BidVol - AskVol) / (BidVol + AskVol)
Positive = bullish pressure
1.1 How Polymarket Works
Understanding the basics of prediction market trading
By the end of this lesson, you'll understand what prediction markets are, how Polymarket facilitates trading, and the fundamental mechanics of buying and selling shares.
What is a Prediction Market?
A prediction market is a marketplace where participants trade contracts whose payoffs are tied to the outcome of future events. Unlike traditional betting, prediction markets aggregate information from many participants, making prices reflect collective probability estimates.
Core Concept: Price = Probability
In a prediction market, the price of a share represents the market's estimated probability of an event occurring.
If "Bitcoin > $100K by March 2026" trades at $0.45, the market believes there's a 45% chance this will happen.
- If it happens, YES shares pay out $1.00 (profit: $0.55)
- If it doesn't, YES shares pay out $0.00 (loss: $0.45)
How Polymarket Works
Polymarket is a decentralized prediction market built on the Polygon blockchain. Here's the key architecture:
The Trading Flow
Deposit USDC
Fund your wallet with USDC on Polygon. This is your trading capital and serves as collateral for positions.
Find a Market
Browse markets by category (politics, crypto, sports, etc.) or search for specific events you want to trade.
Place Your Order
Buy YES if you think the event will happen, or NO if you think it won't. Set your price and quantity.
Wait for Resolution
The market resolves when the event occurs (or doesn't). The UMA oracle determines the outcome.
Collect Winnings
If your prediction was correct, your shares pay out $1.00 each. Redeem them for USDC.
YES vs NO Shares
Every market has two types of shares that are mathematically linked:
YES Shares
- Pay $1.00 if event happens
- Pay $0.00 if event doesn't happen
- Buy when you think probability is underestimated
NO Shares
- Pay $1.00 if event doesn't happen
- Pay $0.00 if event happens
- Buy when you think probability is overestimated
Key Insight: YES + NO = $1.00
The prices of YES and NO shares always sum to approximately $1.00. If YES trades at $0.60, NO trades at about $0.40. This is because one of them must pay out.
Why Prediction Markets Work
Prediction markets are effective because they:
- Incentivize accuracy - You profit by being right, so you're motivated to find true probabilities
- Aggregate information - Prices reflect the combined knowledge of all participants
- Update in real-time - News and events are immediately reflected in prices
- Resist manipulation - Attempts to manipulate create profit opportunities for others
Practice: Calculate Your Edge
A market shows YES at $0.35. You believe the true probability is 50%.
Question: What's your expected profit per share if you buy YES?
Show Solution
Expected Value = (Probability × Payout) - Cost
EV = (0.50 × $1.00) - $0.35 = $0.50 - $0.35 = $0.15 per share
This represents a 43% expected return on your $0.35 investment!
1.2 CLOB Architecture
Deep dive into the Central Limit Order Book
This section covers technical architecture. Understanding this helps with API integration and building trading systems.
What is a CLOB?
A Central Limit Order Book (CLOB) is a trading system that matches buy and sell orders based on price and time priority. Polymarket uses a hybrid CLOB model:
Off-Chain
- Order matching
- Order book management
- Price discovery
- Low latency execution
On-Chain
- Trade settlement
- Token custody
- Collateral management
- Final state of truth
Order Book Visualization
The order book shows all open buy (bid) and sell (ask) orders:
Order Types
Limit Order
Specify exact price. Only executes at your price or better.
Buy 100 YES @ $0.45
Waits until someone sells at $0.45 or lower
Market Order
Execute immediately at best available price.
Buy 100 YES @ market
Fills immediately at best ask ($0.52)
GTC (Good Till Cancelled)
Order stays open until filled or cancelled.
Buy 100 YES @ $0.40 GTC
Stays in book indefinitely
FOK (Fill or Kill)
Must fill completely immediately or cancel.
Buy 100 YES @ $0.52 FOK
Cancels if can't fill all 100
Order Lifecycle
# Order states in Polymarket CLOB
order_states = {
"LIVE": "Order is active in the order book",
"DELAYED": "Awaiting operator signature",
"MATCHED": "Partially or fully filled",
"CANCELLED": "Cancelled by user or system"
}
# Typical lifecycle
# 1. User signs order (EIP-712)
# 2. Order submitted to CLOB
# 3. Operator validates and countersigns
# 4. Order becomes LIVE
# 5. Matching engine finds counterparty
# 6. Trade settles on-chain
Price-Time Priority
Orders are matched using price-time priority:
- Price Priority: Better prices execute first (higher bids, lower asks)
- Time Priority: Among orders at the same price, earlier orders execute first
Three buy orders arrive:
- Alice: Buy 100 @ $0.50 (arrives 10:00:00)
- Bob: Buy 200 @ $0.51 (arrives 10:00:01)
- Carol: Buy 150 @ $0.50 (arrives 10:00:02)
If someone sells 250 shares at market:
- Bob's order fills first (200 @ $0.51) - best price
- Alice's order partially fills (50 @ $0.50) - same price, earlier time
- Carol's order doesn't fill - same price, later time
Key Metrics
Code: Fetching Order Book
import httpx
async def get_orderbook(token_id: str) -> dict:
"""Fetch current order book for a market."""
url = f"https://clob.polymarket.com/book"
params = {"token_id": token_id}
async with httpx.AsyncClient() as client:
response = await client.get(url, params=params)
return response.json()
# Response structure:
# {
# "bids": [
# {"price": "0.48", "size": "15000"},
# {"price": "0.47", "size": "10000"},
# ...
# ],
# "asks": [
# {"price": "0.52", "size": "15000"},
# {"price": "0.53", "size": "10000"},
# ...
# ]
# }
def calculate_spread(orderbook: dict) -> float:
"""Calculate bid-ask spread percentage."""
best_bid = float(orderbook["bids"][0]["price"])
best_ask = float(orderbook["asks"][0]["price"])
mid = (best_bid + best_ask) / 2
return (best_ask - best_bid) / mid * 100
1.3 CTF & Token Mechanics
Understanding the Conditional Token Framework
1.4 API Essentials
Working with Polymarket APIs
1.5 Fee Structure
Understanding trading costs and rebates
2.1 Market Efficiency
How accurate are prediction markets?
2.2 Liquidity & Microstructure
Understanding order flow and market depth
2.3 Price Discovery
How information gets incorporated into prices
2.4 Arbitrage Opportunities
Finding and exploiting market inefficiencies
3.1 Strategy Overview
Nine proven profit models for prediction markets
3.2 Market Making
Providing liquidity for profit
3.3 Directional Trading
Betting on outcomes with edge
3.4 Copy Trading
Following successful traders
3.5 Professional Traders
Archetypes and case studies
4.1 Probability Calibration
Measuring and improving forecast accuracy
4.2 Kelly Criterion
Optimal position sizing for maximum growth
4.3 Bayesian Methods
Updating beliefs with new information
4.4 ML for Forecasting
Machine learning approaches to prediction
4.5 Superforecasting
Techniques from the world's best forecasters
5.1 Platform Risks
Smart contract, oracle, and operational risks
5.2 Regulatory Landscape
CFTC, international regulations, and compliance
5.3 Tax Implications
Understanding tax treatment of prediction market profits
5.4 Risk Management
Position sizing, stop losses, and portfolio management
Platform Comparison
Polymarket vs Kalshi vs PredictIt vs Manifold
Tools & APIs
Essential tools for prediction market trading
Glossary
Key terms and definitions
External Resources
Curated links for further learning